AgentOps API MCP Server

@AgentOps-AI

Basic information

About AgentOps API MCP Server

Official MCP server for the AgentOps SDK

AgentOps API MCP Server config

Add this server to your MCP-compatible client using the configuration below.

{
  "mcpServers": {
    "agentops-mcp-server": {
      "command": "uv",
      "args": [
        "run",
        "-m",
        "mcp_server_agentops_api"
      ]
    }
  }
}

AgentOps API MCP Server tools

Tool list not available yet

This listing does not currently have a tool list to display. This does not mean the server has no tools; check its documentation for supported capabilities.

We auto-extract tools from the README. The maintainer can list them under a ## Tools heading to populate this section.

Overview

What is AgentOps API MCP Server?

An MCP server that enables LLM agents to interact with the AgentOps API to access trace data. It provides tools for listing recent traces and retrieving detailed information about specific traces, handling authentication and API communication.

How to use AgentOps API MCP Server?

Configure it as an MCP server in Claude Coder (e.g., in ~/.config/claude-cli/config.yaml) by specifying the server binary path, then run Claude with the --mcp agentops-api flag and include your AgentOps API key in prompts. For standalone use, run ./bin/run-server or uv run -m mcp_server_agentops_api.

Key features of AgentOps API MCP Server

List recent traces from an AgentOps project
Get detailed information about specific traces
Secure two-step authentication (API key → JWT token)
Optional custom API URL for non-production environments
Stateless design – no user data stored between requests

Use cases of AgentOps API MCP Server

An LLM agent asks for the most recent traces in a project
An agent drills into a specific trace to inspect spans and metadata
Debugging agent workflows by fetching trace timing and details
Integrating trace data retrieval into automated agent pipelines

Frequently asked questions

What tools does the server provide?

Two tools: `list_traces` (with parameters `AGENTOPS_API_KEY`, `limit`, `AGENTOPS_API_URL`) and `trace_detail` (with parameters `AGENTOPS_API_KEY`, `trace_id`, `AGENTOPS_API_URL`).

How does authentication work?

The server requires an AgentOps API key with each request. It exchanges the key for a JWT token, which is used for all authenticated API calls. API keys are never stored between requests.

Where does trace data live?

Data resides in the AgentOps API (production at `https://api.agentops.ai` by default). The server does not store any data – responses are passed directly to the agent.

What are the runtime dependencies?

The server uses Python and can be run via the provided script (`./bin/run-server`) or with `uv run -m mcp_server_agentops_api`. A Docker image can also be built and run.

Can I use a custom API URL?

Yes, both tools accept an optional `AGENTOPS_API_URL` parameter to override the default production endpoint for local development or staging environments.

Comments

More Developer Tools MCP servers

OrangePro

Aamir Siddiqui

OrangePro local-first CLI + MCP server for behavior mapping, grounded test generation, and dynamic proof.

Mcp-Warden

ermoluk

Local-first credential broker for MCP. AI coding agents (Claude Code, Cursor, Codex…) call APIs, SSH and databases through scoped presets; secrets stay in your OS keychain and never enter the agent's context.

Vibgrate MCP

Vibgrate

Query your team's drift, vulnerability, and migration data from any AI assistant. Vibgrate MCP connects Cursor, Claude, ChatGPT, Windsurf, or VS Code to Vibgrate Cloud: 51 tools for DriftScores, CVEs and EOL runtimes, up

Modelglass

Modelglass turns "which model should I use?" into a lookup instead of a guess. Seven tools give your agent live pricing across image, language, video, and audio models. Sourced and versioned, not scraped once and forgott

Ybug

Connect Claude, ChatGPT, Cursor or any MCP client to Ybug. Read bug reports with screenshots, console logs and network requests, then triage them in chat.